US3364382A

US3364382A - Automatic generation and display of animated figures

Info

Publication number: US3364382A
Application number: US607078A
Authority: US
Inventors: Iii Lee Harrison
Original assignee: Control Image Corp
Current assignee: Control Image Corp
Priority date: 1967-01-03
Filing date: 1967-01-03
Publication date: 1968-01-16
Anticipated expiration: 1985-01-16

Description

AUTOMATIC GENERATION AND DISPLAY OF ANIMATED FIGURESv l2 Sheets-Sheet l Original Filed Nov. 29, 1962 Jan. 16, 1968 L.. HARRISON lu AUTOMATIC GENERATION AND DISPLAY OF ANIMATED FIGURES Original Filed Nov. 29, 1962 Jan. 16, 1968 HARRISON 3,364,382

AUTOMATIC GENERATION AND DISPLAY OF ANIMA-TED FIGURES Original Filed Nov. 29, 1962 l2 Sheets-Sheet 5 Jan. 16, 1968 L HARRlsoN nl AUTOMATIC GENERATION AND DISPLAY OF' ANIMATE!) FIGURES l2 Sheets-Sheet 4 Original Filed Nov. 29

w a WW w00 n l w MW n, m f N R MJ m A m H E @www 55 L Rw r w u gu. SG

@uw mm TTORA/EYS Jan. 16, 1968 L HARRISON lll AUTOMATIC GENERATION AND DISPLAY OF ANIMATED FIGURES l2 Sheets-Sheet 5 Original Filed Nov. 29, 1962 mmm mmm

.d llll 325m anw WFT@

vxy

mmm

Qms

Jan. 16, 1968 L. HARRISON nl 3,364,382

AUTOMATIC GENERATION AND DISPLAY OF ANIMATED FIGURES Original Filed Nov. 29, 1962 l2 Sheets-Sheet 6 569 492 435 //Hr 420 "if," 438 LLB-N 437 427 442 2 33315 L 327 567 5mg;

R/GHT l 413 LOT 540E 509 440 .421

\ 570 456 414 PL-f ,-4 575555 WL 3 6 V 462 f STKE? 440 429 475 fes `1 44g maar 367 I 452 424 43] 4,7 geef-Mr 449 m LEFT s'rArer 432 418 LEFT f SIUE 545 543 426 557 555 553 433 aLg/@6447' srUE 7T ATTO RMD/5 Jan. 16, 1968 1 HARRISON 111 AUTOMATIC GENERATION AND DISPLAY OF ANIMATED FIGURES 12 Sheets-Sheet Original Filed Nov. 29

Jan. 16, 1968 L. HARRISON m AUTOMATIC GENERATION AND DISPLAY OF ANIMATED FIGURES 12 Sheets-Sheet 8 Original Filed Nov. 29

QTTORNEKS Jan. 16, 1968 1 HARRISON m 3,364,382

AUTOMATIC GENERATION AND DI SPLAY OF ANIMATED FIGURES Original Filed Nov. 29, 1962 l2 Sheets-Sheet mYaNTOR LEE HARRIS-@Nil Q TTORNE Y6 Jan. 16, 1968 HARRISON nl 3,364,382

AUTOMATIC GENERATION AND DISPLAY OF ANIMATED FIGURES Original Filed Nov. 29, 1962 l2 Sheets-Sheet lO Lcosdbsn/Q p Hs//wpcosKzt w.

INVENTOR HCoSM/acos/(zt LEE HRRSON EI HTTOENE Y Jan. 16, 1968 L.. HARRISON 3,364,382

AUTOMATIC GENERATION AND DISPLAY OF ANIMATED FIGURES Original Filed Nov. 29, 1962 Y l2 Sheets-Sheet 11 BMW/wlw #mw Jan. 16, 1968 L. HARRISON lll AUTOMATIC GENERATION AND DISPLAY OF ANIMATEDFIGURES l2 Sheets-Sheet l2 Original Filed Nov. 29

QQQ

Q N5 QHHQ M MWL a Q Q u E Z ATTo/QNEV United States Patent O 3,364,382 AUTMATIC GENERATION AND DISPLAY Oli? ANIMA'IED FIGURES Lee Harrison Ill, Norristown, Pa., assignor to 'Control Image Corporation, New York, NX., a corporation of Delaware Continuation of application Ser. No. 240,970, Nov. 29, 1962. This application Jan. 3, 1967, Ser. No. 607,078 53 Claims. (Cl. SiS- 18) This is a continuation of Ser. No. 240,970, now abandoned.

This invention relates to a system for generating one or more figures, animating the figures, and displaying the animated figures as a series of high frequency displays. The general object of the invention is to provide a system whereby an operator can regulate a small number of inputs to generate one or more animated three dimensional figures which are thereafter resolved into two dimensions to produce an animated display on a display tube.

Broadly speaking, this invention provides a system for generating and displaying a sequence of picture frames at a frame rate which is compatible with the object of the display. If the display is for transmission over television, the frame rate would be identical to television frame rate, or if the display is to be photographed, the frame rate would correspond to that for motion picture photography. At any rate, the ultimate sequence of display can accommodate any motion of the display subjects including motions of human figures, cartoons and moving objects.

The subject matter to be displayed is st-ored information available to the system. This subject matter is animated by operation of the variable inputs to the machine, these inputs being any variable transducing elements, such as potentiometers or capacitors. These variable inputs are in circuits which relate to the solution of parametric equations to locate the different parts of the subject matter in three dimensions. As such, the variable inputs may be hand or mechanically operated controls, or they may be designed to receive variable signals from potentiometers or capacitors connected directly to movable members of a physical body for transmitting signals which vary in proportion to angular movements of the movable members. Whatever the input, animation can be created by an operator and the displayed figure can be made to go through all movements imaginable. In the case of live figure input, the system can be made to reproduce movements of the ligure even though the figure be many miles distant from the system.

The principal components of this system include a master oscillator or clock, circuitry for generating voltages representing the axes of the different members of the figures and/or objects to be animated, hereinafter referred to as a bone generator network, and circuitry for generating voltages representing the radial distances of points on the surfaces of the figures and objects from their respective axes, hereinatfer referred to as a skin generator network. The clock controls the operation of the bone and skin generator networks. The bone generator network includes a means for generating groups of pulses for durations representing the lengths of various axes of members of figures and objects, conveniently called bones. At the same time, various voltages are introduced to position these bones in three-dimensional space. The positioning voltages are treated by a network that generates various trigonometric functions of the voltages which are parts of different parametric equations which must be solved to determine the differentpositions of different members being drawn. These trigonometric functions are then transmitted to an integrator the output of which produces voltages representing the instantaneous value of the bone positions.

3,364,382 Patented Jan. 16, 1968 The skin generator network has a means for scanning stored information to modulate the magnitude of a variable skin vector according to the distance of the skin from the bone. This variable length vector is treated by a network that superposes the trigonometric functions of the positioning voltages to relate the skin vector to the proper bone, and thereafter the skin vector is added to the bone.

The three-dimensional figure thus generated is transmitted to a camera angle network that can select any viewing angle and can transpose the three dimensions viewed from that angle into a two-dimensional display on the face of the display tube.

An important object of the invention is to provide a system that permits an operator to establish the levels of a plurality of variable inputs according to his desired animation pattern and that provides for recording inputs for automatic regulation of the system upon playback of the recorder to produce an automatic animated display on the face of a display tube.

Another object of the invention is to provide a system for generating and displaying animated sequences of one or more figures with provisions for controlling the variable inputs to generate and animate the figures automatically by stored information.

With the foregoing objects in mind, it is an object of this invention to provide a fast, lower cost means of picture animation with such a broad range of control and automation that the artistic range of the system is limited only by the operators imagination.

It is another object of the invention to provide automatic display of the motion of a figure wherein the generation of motion in the system is produced by changes in low frequency, low bandwidth inputs so that the information dictatng changes in these low bandwidth inputs can ybe transmitted over communications means of low bandwidth capabilities.

A more specific object of the invention is to provide a system having a network for generating the bones of a figure, a network for generating the skin associated with those bones, and a network for adding the skin to the bones to produce a three-dimensional figure, and also having a network for viewing the three-dimensional figure from any angle and displaying the figure as thus viewed. An additional object is to provide means for animating the figure,

Another specitic object of the invention is to provide means for generating and animating a figure for display by the successive generation of the physical members of the figure with means to prevent overlap of the display when the generation of more than one of the physical members takes place at least in part over the same area of the display.

Still another specific object of the invention is to provide a system for generating and displaying animated figures and for modulating the intensity of the display to incorporate the minute physical characteristics of the tigure and to provide shading for the figure.

Other objects and advantages will be apparent to those skilled in the art.

In the drawing:

FIGURE l is a block and schematic diagram of the clock, integrator, and flyback networks;

FIGURE 2 is a block diagram of the step counters and bone gates;

FIGURE 3 is a block and schematic diagram of the sine-cosine function generator;

FIGURE 4 is a block and schematic diagram of the equation solving network;

FIGURE 5 is a block and schematic diagram of the camera angle network and the gross position network;

FIGURE 6 is a block and schematic diagram of the program network for the skin scanning network;

FIGURE 7 is a block and schematic diagram of the skin scanning network;

FIGURE 8 is a block and schematic diagram of the display tube, the overlap prevention network, and the background information generator;

FIGURE 9 is a plan view of a typical skin iilm;

FIGURES 10-14 are geometric diagrams illustrating the general theory of bone and skin generation;

FIGURE 15 is a geometric diagram illustrating the theory of generation of bones and skin for Mode Two operation;

FIGURES 16-18 are geometric diagrams illustrating the general theory of the camera angle network;

FIGURE 19 shows a typical gure display in Mode One operation;

FIGURE 2O shows a typical iigure display in Mode Two operation; and

FIGURE 21 is a block and schematic diagram of the recording network.

General theory .and analytical geometry of bone and ski/z generation for lllode 011e or full guie operation For purposes of illustration, the emphasis throughout the description of this invention centers upon the drawing of a human figure or animated figure having physical limbs and members. The text describes how the gure is generated for display on the face of a display tube, explaining the generation of each series of bones for the various parts of the gure, the generation of skin added to these bones, and the animation of the figure. The immediate investigation concerns the general theory of bone and skin generation, and for this consideration a typical bone and the skin for that bone are analyzed geometrically, as illustrated in FIGURES 10-14.

A typical bone is designated L in FIGURE 10, The bone is drawn at a constant rate of speed so that its length depends upon the rate that the drawing beam moves in drawing the bone and the period of time during which the drawing occurs. The rate is a constant for a given mode of operation and may be designated k1. The time is variable and is designated t. Therefore, the length of the bone L is klt.

The bone L is a single straight line as shown in FIG- URE 10. Skin is added to the bone by what may be regarded as a twirling vector A that continues to rotate about the bone L. The vector A moves from the start of the bone to the end of the bone during the period of time t. As the bone L is generated, and during each incremental portion of the time t, the vector A rotates 360 about the bone L. (These increments of time t will be more readily understood hereinafter.) A typical revolution of the end of the vector A is indicated in dotted lines P on FIGURE 10. The drawing ultimately made on the display tube depends upon the position of the tip or end of this vector A.

As viewed in FIGURES 10 and 11,'the vector A may be thought of as rotating in a clockwise direction. It rotates at a constant angular speed the rate of which may be designated K2. Therefore, the angular position of the vector A depends upon the product Kzz.

The generation of the bone L will be considered rst. Since the information available for display consists of voltages representing a three-dimensional gure, FIG- URE 10 shows the bone L in reference to three-dimensional axes X, Y and Z. The angle that the projection of the bone L in the X, Y plane makes with the X axis is designated 0. The angle that the bone L makes with the bone L makes with the X, Y plane is designated qb. An examination of FIGURE 10 reveals the X, Y and Z components of the bone L.

The length of the projection of the bone L on the X, Y plane is equal to L cos qb. Therefore, X:L cos qb cos 0. But since the length of the bone L is kll, x=k1z cos 0 cos q).

It follows that Y-Jclr sin 0 cos qb and Zzkll sin qb.

In considering the generation of skin for the bone L,

it may be assumed that the vector A always rotates in a plane perpendicular to the bone, although this angle may be varied. In FIGURES 10-13, the plane of rotation of the vector A is drawn perpendicular to the bone L. As already mentioned, the angular position of the vector A depends upon the product k2t. FIGURE 1l shows this plane of rotation of the vector A and is drawn perpendicular to the bone L. As shown in FIGURE l1, the vector A always has two components that vary with the cosine and sine of the angle kzz. The length of these components are A cos 1:21* and A sin kzl. These components are shown on FIGURE 11 with the appropriate legends.

FIGURES 12, 13 and 14 show how the skin vector A is resolved into its X, Y and Z components. The coordinates of FIGURE 12 are the Z axis and the X, Y plane and the plane of FIGURE 12 is defined by the Z axis, the bone L, and the projection of the bone L on the X, Y plane, that projection bearing the legend L cos qb in FIGURE 10. Since FIGURE l2 shows the bone L in its true length, it views the plane of rotation of the vector A from the side. Therefore, that plane, designated P, appears as a straight line in FIGURE 12, normal to the `bone L, and with the length above the bone L and the length below the bone L each being equal to A cos k2t. Since the angle between the plane P and a vertical line drawn from the end of the bone L is equal to qb, a horizon-tal line connecting that vertical line with the end of the plane P is equal to A sin qb cos k2t. In other words, the projection of the A cos kzt vector on the X,Y plane abscissa of FIGURE 12 equals A sin qb cos kzt. The projection of this A cos kzt vector on hte Z axis equals A cos qb cos kzz, which is the Z component of the A vector, since in FIGURE 12, A sin k2=0.

FIGURE 13 is a plane through the X and Y coordinates projected from FIGURE 12. In this view, the maximum length of a line drawn from the end of the L cos qb projection to the outer extremity of the plane P is equal to the A sin kgz component of the vector A. Since the angle between the L cos qb projection and the X axis is 0, the component A sin k2t can be resolved into its X and Y components as indicated, whereby the X component is A sin H sin kzt and the Y component is A cos 0 sin k2t.

In FIGURE 13, the projection A sin qb cos kzt is also shown, and as illustrated in FIGURE 14, this component may be resolved into X and Y components whereby the X component is A cos 0 sin qb cos kgt, and the Y component is A sin 0 sin qb cos kgt.

Since the vector A rotates 360 about the bone L, its X, Y and Z components will vary between positive and negative values. However, from an examination of the direction of the 4vectors illustrated in FIGURES 13 and 14, it can rbe seen that the net X component of the vector A is always equal to the difference between the quantities A cos 0 sin qb cos kzt and A sin 0 sin kzt, and the Y component of the vector A is always equal to the sum of the components A sin 0 sin qu cos kzt and A cos 0 sin kzt.

From the foregoing description, it is evident that the components for the generation of the bone L with skin are as follows:

X=k1t cos 6 cos qb-I-A cos 6 sin qb cos kZt-A sin H sin k2t Yzklz sin 0 cos qi-j-A sin 0 sin qb cos kZt-l-A cos 0 sin kg Z=k1t sin qb-j-A cos qb cos kzt Bone and skin generation for mode Iwo or )gure outline As will be explained hereinafter, there are times, especially during rapid animation, when only an outline of the figure is to be drawn. For Mode Two operation, the gure displayed on the face of the display tube shows only an outline of the skin in the X, Y plane.

For Mode Two, an appropriate constant is substituted for the high frequency sinusoidal factors sin kzt and cos k2t of the general equations for X, Y, and Z. These equations then represent the generation of skin volume. In other words the twirling vector A is no longer twirling, and the specific case of interest is the solution to the,

general equations when legt-:90 and k2t=-90, that is, when the vector A is parallel to the viewing plane, for this case the XY plane. (The selection of i90 for angle kgz is in keeping with the phase coordinates of vector A which were used to develop the general equations. In other words (FIGURE 13) when k2t=90, A sin k2t=A 1:14.)

Therefore, by substituting the values i90 for k2! in the X, Y and Z components of the general equations, the resulting equations are:

In this particular case there is no Z component of A, and the ligure being generated may 'be thought of as being bat.

FIGURE l5 illustrates the theory of skin and bone generation for Mode Two. In Mode Two, the generation of the X, Y and Z components for the bone L is the same as was described in connection with FIGURES -14. To add the X, Y and Z components of the skin outline, it is only necessary to determine the X, Y and Z components of the circumference of the circle P generated by rotation of the vector A. The radius of this circle P is A. Referring to FIGURE 15, it is evident that the projection of the vector A in the XY plane does not change the Ylength of the vector A. Since the angle that this projection A of the vector A makes with a line D drawn normal to the axis A is 0, it follows that the X component of the vector A is A sin 9 and the Y component is A cos 6.

Accordingly, the equations for the X, Y and Z components of the bone and skin are as follows:

In these equations, the A component may be positive or negative.

Geometric theory of camera angle network-resolution into' two dimensions The three-dimensional ligure must be resolved into horizontal H and vertical V components for display on the face of the display tube. To do this, the three components X, Y and Z of the three-dimensional ligure must be resolved into two components H and V.

To illustrate this resolution, it may be assumed that the X, Y and Z axes of FIGURE 16 represent the X, Y and Z components of `a point that is to be resolved into two components. As the system is illustrated, the entire figure is rotatable about the Z axis. The angle through which this rotation occurs is designated a as indicated in FIG- URE 17. This rotation produces new coordinates X 'and Y wherein X=X cos ar-i-Y sin a and Y.=Y cos a-X sin a The system also provides for rotation of the Y Z plane about the X axis, as illustrated in FIGURE 18. This angle of rotation is designated b and establishes two axes Y and Z. This rotation of FIGURE 18 produces the quantities Y=Y' cos b-Z sin b and From an analysis of FIGURES 17 and 18, it is apparent that the quantities X and Y may be used to represent two dimensional axes wherein variations of the angles a and b permit viewing of a three-dimensional figure from any angle. Therefore, the components of -the display scope are as follows:

H=X cos a-l-Y sin a V=(Y cos a-X sin a) cos b--Z sin b ,Clock control Referring to FIGURE l, the entire system is regulated and controlled by a 'high frequency master oscillator 1t).

d A cathode ray display tube 11, shown in FIGURE 8, develops a display that can be photographed. The network between the master oscillator or clock 10 and the display tube 11 determine the nature of the display.

The clock 1l) has two

wave outputs

12 and 14. The frequencies of the waves at the

outputs

12 and 14 are the same, but one of the outputs 12 is a square wave and the other one 14 is a conventional sine wave. The sine wave 14 generated by the master oscillator is itself used in the system, and this sine wave output is also fed through a phase shifter 15 the output 16 of which is a cosine wave 90 out of phase with, but of the same frequency as, the sine wave output 14. The functions of the sine wave output 14 and the cosine wave output 16 will be described hereinafter.

The square wave output 12 is carried to the successive inputs of a series of

bistable multivibrators

18, 19, 26, 21, 22 and 23N, each of which halves the frequency of its input. (Here and elsewhere in this description the suix N is used to indicate there may be a variation in the uumber of devices used.) Each of the bistable multivibrators has an output 2.4, 2S, 26, 27, 28 and 29N, respectively, which, except for the last output 29N, is connected to the input of the next succeeding multivibrator. An appropriate number of such multivibrators IS-ZSNA are used, so that the frequency of the output of the last multivibrator 23N is equal to an acceptable frame frequency ultimately used for the display to be photographed. For example, an acceptable and conventional frame frequency for motion picture films is twenty-four frames per second. Therefore, if six bistable multivibrators 18-23N a-re used, the frequency of the square wave 12 (and the sine wave 14) generated by the master oscillator is set at 1536 cycles per second. Accordingly, changes in the output frequency of the master oscillator produce changes in the frame frequency unless additional multivibrators 11S-23N are used. Since the master oscillator 10 regulates everything else in the system, any such change in its frequency also causes other system operations to remain synchronized with the frame frequency. As will be evident hereinafter, the higher the frequency of the master oscillator 10, the greater will be `the resolution of the final picture displayed. A higher frequency oscillator merely requires the use of additional multivibrators liti-23N.

The outputs 24-29N from them ultivibrators are also delivered through

cathode followers

30, 31, 32, 33, 34 and 35N, respectively (or through 'buffer amplifiers or similar devices to stabilize the back impedance as is convention-al in the art), to individual terminal plugs 36, 37, 38, 39, 40 and 41N, respectively. It is possible that during operation of the system not all of the bistable multivibrator outputs 2.4-29N will be used, except as an input to lthe next multivibrator, but the last multivibrator 23N always feeds its frame frequency square wave output 29N through a conductor ft2 to the first of a series of storage counters or step counters 46, 47, 48, 49 and 50N (see FIGURE 2). The lframe pulse output 29N is also used for other purposes that will be described.

Bone generators inputs is connected to one of the terminal plugs 36, 37, 33,

39, 40 or 41N. Although not always, the inputs 52-56N are sometimes connected to a common terminal plug, such as to the plug 37 as indicated by dotted lines on the drawing. The length of the bone being drawn, and the capacity of the storage counter determines the choice of connections 36-41N. Each storage counter counts a variable number of pulses transmitted to its

inputs

52, 53, 54, 55 or 56N.

The duration of the set state of each storage counter is controlled by an intrinsic capacitive network (not shown) wherein the capacitor is variable to provide independent regulation of the set state for each storage counter. These variable capacitors may be controlled by conventional hand controls 37, 58, 59, 6d and 61N associated with the storage counters 465914, respectively. The setting of a variable capacitor, such as the control 57, determines the number of pulses presented to the input 52 that the storage counter 46 will count.

Although only five storage counters 46-5tlN are illustrated, there are actually a much larger number. The storage counters are in convenient groups of various number depending upon what object they are associated with. For example, if a human figure is to be drawn, there may be four storage counters 46, 47, 43 and 49 for serially stepping off lengths of a placement bone, the upper arm bone, the lower arm bone, and the hand. For purposes of illustration, the four storage counters 46, 47, 48 and 49 constitute such an arm group and the storage counter 50N may be thought of as the first of a series constituting another group, as a leg group.

The rst storage counter 46 is triggered by the frame pulse 29N (see FIGURE l) transmitted through the conductor 42 and ips to its set state for a duration determined jointly by its input 52 and the control 57. The storage counter 46 has an output 67 the Voltage level of which changes when the storage counter changes states. This change in the output voltage 67 is fed through a cathode follower 63 (or buffer amplifier) and provides a common (operating) input to a blank of

gates

69, 70, 71 and 72 to open the gates for the period of time the storage counter 46 is in its set or pulse counting state.

The step counter 46 automatically flips back to its reset or quiescent state at the end of the period determined by the input 52 and the control 57. At this time, the storage counter 46 delivers a voltage to another output 74, which voltage is of the correct value to flip the next storage counter 47 to its set state. For the duration of the set state of the storage counter 47, which is determined by its input 53 and the control 53, a change in voltage at an output 75 occurs which is fed through a cathode follower 76 and simultaneously opens a bank of gates 77, '73, 79 and 8). Upon flipping back to its reset state, the storage counter 47 generates a voltage at another output 82 that is of proper value to ip the next storage couner 48.

The storage counters 43 and 49 are connected to operate like the storage counters already described. Thus, the storage counter 4S has an output 83 fed through a cathode follower 84 that opens a bank of

gates

85, 86, 87 and 38 during the set state and an output 90 that causes the next storage counter 49 to flip to its set state. The storage counter 49 has an output 91 that goes through a cathode follower 92 and opens a bank of

gates

93, 94, 95 and 96 and an output 93 that flips the next storage counter. However, the storage counter 49 is lthe last one of the arm group, which leads to the significance of the and gates and or gates.

In the preceding description, it was assumed that the storage counters 46-49 were directly connected together in a series chain. Actually, the input pulse 42 to the first storage counter must first pass through an or gate 110. The output 74 from the storage counter 46 must pass through an and gate 111 and an or gate 112 before it can trigger the storage counter 47. The output 32 from the step counter 47 must pass through an and gate 113 and an or gate 114 before it can trigger the storage counter 43. And the output 919 from the storage counter 43 must pass through an and gate 115 and an or gate 116 before it can trigger the storage counter 49. Also, the output 93 from the last storage counter 3 49 of the arm group is delivered as an input to an and gate 117.

There is an in-out bistable multivibrator 120 having an out input conductor 121 connected to the output conductor 42 from the frame pulse multivibrator 23N. Therefore, when a trigger pulse is transmitted to the or gate 110, it is also delivered to the multivibrator 121) and flips the multivibrator to its out condition. The multivibrator 121i has an out output 122 that passes a volta-ge when the multivibrator is in the out condition. This output is delivered as

inputs

123, 124, 125 and 126 to the and

gates

111, 113, 115 and 117, respectively.

The in-out multivibrator 120 also has an in input 123 connected to the output from the and gate 117 on the output side of the storage counter 49. A signal in the in input 123 flips the multivibrator 120 to its in condition. There is an in output conductor 129 that receives a voltage when the multivibrator is in its in condition. This conductor simultaneously delivers whatever Voltage it carries to a group of and

gates

130, 131, 132 and 133.

Another input conductor 134 to the and gate 13G is connected from the output side of the storage counter 49. An input conductor 135 to the and gate 131 is connected from the output side of the step counter 48. An input conductor 136 to the and gate 132 is connected from the output side of the step counter 47. And an input conductor 137 to the and gate 133 is connected from the output of the step counter 46.

The and gate 117 has an output conductor 138 connected as an input to the or gate 116 on the input side of the step counter 49. The and gate 130 has an output conductor 139 connected as an input to the or7 gate 114. The and gate 131 has an output conductor 140 connected to the input side of the or gate 112. The and gate 132 has an output conductor 141 connected to the input side of the or gate 110.

The and gate 133 has an output conductor 142 connected to the input side of another or gate 143 leading to the first step counter 66N of the next (leg) group of step counters. This step counter 50N has an output 145 that is connected through a cathode follower 146 to a bank of

gates

147N, 148N, 149N and 153N.

The several and gates and or gates just described are of conventional construction. Each and gate transmits an output signal only when there are simultaneous inputs at both its inputs. Each or gate acts as a valve that will pass a voltage at either of its inputs to its output, but not to the other input.

With the bistable multivibrator 120 hooked up as described, it is ilipped to its out condition Whenever a frame pulse from the last multivibrator 23N passes through the conductor 121. While the multivibrator 120 is in its out condition, it passes a voltage through the out output 122. At this time, there is no signal in the in output conductor 129. Hence the and

gates

130, 131, 132 and 133 pass no signal through their

output conductors

139, 140, 141 and 142 to the or

gates

114, 112 and 110 and 143. Under these conditions, the signal from the conductor 42 can pass through the or gate 110 to the step counter 46. Since the out conductor 122 is delivering a voltage to the and gate 111, when an output voltage from the step counter 46 reaches the and gate 111 it passes through to the or gate 112 and thence to the step counter 47. Likewise, the output 32 from the step counter 47 passes through the and gate 113 and the or gate 114 to the step counter 48, and the output 9@ from the step counter 48 passes through the and gate 115 and the or gate 116 to the step counter 49.

When the storage counter 49 delivers a voltage to its output 98, that voltage passes through the and gate 117 to its output conductor 133 and also through the in input conductor 128 to the in-out bistable multivibrator 120. This flips the multivibrator 120 to its in condition,

blocking off the out output 122 and causing the trans` mission of a voltage through the in output conductor 129. Now the conductor 122 is delivering no input voltage to the and

gates

111, 113, 115 and 117 so these gates cannot pass any voltages from the storage counters; but the conductor 129 transmits its voltage as inputs to the and

gates

131i, 131, 132 and 133.

Under these conditions, the output voltage 138 passes through the or gate 116 and tlips the storage counter 49 to its set state. When the storage counter 49 ilips back to its reset state, its output voltage 98 cannot pass through the and gate 117, but is provides a second input to the and gate 130 and passes through the conductor 139 to the or gate 114. The or gate 114 passes this voltage and triggers the storage counter 48.

The output 90 from the storage counter 48 cannot pass through the and gate 115, but does pass through the and gate 131, the conductor 140, and the or gate 112 to trigger the storage counter 47. The output 82 from the storage counter 47 passes through the and gate 132 and the or gate 118 to trigger the iirst storage counter 46. Then the output 74 from the storage counter 46 passes through the and gate 133 and the conductor 142 to the or gate 143 and the iirst step counter 50N of the next (leg) group.

Of course each time these storage counters are iiipped to their set states, they cause those gates which are connected to their outptus to open as has been described. Therefore, during the in7 condition of the in-out bistable multivibrator 120, there is an exact reversal in the order of operation of the storage counters and their associated gates.

The iirst gate of each bank is a 6 gate. Thus, the

gates

69, 77, 85, 93 and 147N are 6 gates connected to the successive outputs of the storage counters 46, 47, 48, 49 and 58N as has been described. These 9 gates have variable DC (or other)

inputs

160, 161, 162, 163 and 164N, the magnitudes of which may be independently regulated by hand controlled potentiometers or any number of other means. These DC voltage inputs are passed to the

respective gate outputs

166, 167, 16S, 169 and 170N, all of which outputs are connected to a common conductor 171.

The

next gates

70, 78, 86, 94 and 148N are the tp gates. These gates have variable DC voltage or

other inputs

173, 174, 175, 176, and 177N, which may also be regulated by hand controlled potentiometers. These fp gates have

outputs

178, 179, 181B, 181, and 182N which are connected to a common conductor 183.

The

gates

71, 79, 87, 95 and 149N are r gates for establishing certain rotational conditions, and the

gates

72, 80, 88, 96, and '0N are i gates for regulating the intensity of the display beam. These gates and this function will be described in detail hereinafter.

Sine-cosine function generator The conductor 171 which carries a voltage representing the magnitude of the angle 0 for whatever bone is being drawn is connected-through a resistor 186 to the input side of an operational ampliiier 187 (see FIGURE 3). Another input conductor 188 through a resistor 189 to the operational amplifier 187 comes from the in-out bistable multivibrator 120. A pulse in the conductor 188 operates to shift the angle 6 180 during the in condition of the multivibrator 120. Therefore, the output 190 from the operational amplifier 187 represents either the angle 0 (during the out condition of the multivibrator 120) or a 180 inversion of the angle 0 (during the in condition of the multivibrator 120).

The conductor 183 that carries the outputs from the qb gates 173 through 17'7N is connected through a resistor 192 to an operational amplifier 193. These is another input conductor 194 to the operational amplifier 193 that is connected from the bistable multivibrator 12.0 to shift the angle qb by 180 when the multivibrator 120 is in its in condition. Hence the operational amplifier 193 has an CFI It is practical to perform operations on the quantity @+o and 0- p, thereby eliminating various multiplication steps which are more expensive to do electronically.

To obtain the quantity H-l-go, the outputs and 19S from the operational ampliers 187 and 193 are delivered through a pair of

conductors

197 and 198, respectively, to an operational ampliiier 199 hooked up as an adder. The voltage at the output 280 from the operational amplifier 199 represents the quantity 0-l- ,o.

These

same voltages

190 and 195 are delivered through another pair of conductors 202 and 283 to another operational arnpliiier 204 connected as a subtractor and having an output 205 representing the quantity 6- p.

The output 2011 from the operational amplifier 199 is a constant DC voltage (during the generation of a straight bone) that is fed `through a monostable delay multivibrator 207. The delay multivibrator 207 has another input 208 which is connected to the output of the rst bistable multivibrator 18 on the output side of the master oscillator 19. Therefore, the input 208 to the delay multivibrator is a square wave synchronized with, but at onehalf, the frcquency ot the output of the master oscillator 1li.

The start of the square Wave pulse at the input 208 i'lips the delay multivibrator 207 to its quasi-stable state. The duration of this quasi-stable state is determined by the DC voltage at the input Z011 and is therefore determined by the magnitude of the quantity H-i-p.

Since the input 208 is taken at the output of the rst bistable multivibrator 18 which is one-half the frequency of the master oscillator 18, a single square wave pulse occurs. at the input 2118 during the period of two complete sine waves at the output of the master oscillator 10. Therefore, during this period of time, the sine wave output 14 from the master oscillator 10 goes through the cycle of representing the sine of the angles 0 through 360 twice. This multiple may vary if greater range is desired. The voltage in the conductor 2011 representing the quantity 0-{p determines the duration that the monostable multivibrator 2G17 is in its quasi-stable state. The output 209 from the delay multivibrator is differentiated and clipped to produce a narrow pulse representing the change of state from quasi-stable to stable condition. When the multivibrator 2117 iiips back to its stable state, its output 2819 which is connected to the input of a monostable multivibrator 210 causes the multivibrator 210 to generate an extremely narrow pulse at its output 211. This narrow pulse at the output 211 thus occurs at a time reference to the clock sine wave that is directly related to the magnitude of the quantity f-i-(p put into the delay multivibrator 207.

The narrowpulse-carrying conductor 211 is connected to open two

sampler gates

214 and 215 for the very short period of time corresponding to the length of the narrow pulse. The input to the sampler gate 214 ijs the conductor 14 carrying the sine wave output from the master oscillator 10, and the input to the sampler gate 215 is the conductor 16 carrying the cosine Wave output from the 90 phase shifter 15 on the output side of the master oscillator 11i. Therefore, each time the narrow pulse occurs in the conductor 211, a small portion of the sine wave is sampled by the gate 21d, and a small portion of the cosine wave is sampled by the gate 215. Since the sine and

cosine waves

14 and 16, respectively, are synchronized with the square aserssz wave input 208 to the delay multivibrator 267, and the delay voltage 26@ is proportional to -l-e, the sine and cosine waves sampled represent the sine and cosine, respectively, of the quantity p.

The output conductor 216 from the sampler gate 214 delivers its voltage to a holding capacitor 217 and the output conductor 218 from the sampler gate 215 delivers its output to a holding capacitor 219.

The course of the voltage in the conductor 265 representing the quantity 0 p will now be apparent. This voltage is delivered to a delay monostable multivibrator 222 having the same square wave input 2138 that is the input to the previously discussed delay multivibrator 267. The magnitude of the voltage 205 representing the quantity 0 p determines the duration of the unstable state of the delay multivibrator 222 and the output 223 from the delay multivibrator 222, which occurs when .the multivibrator ips from its unstable state back to its stable state, triggers a narrow pulse generator 224. The narrow pulse output 225 from the multivibrator 224 opens a pair of sampler gates 226 and 227, the inputs to which are the sine wave 14 and the cosine wave 16 from the master oscillator 1t). The quick sampling of these sine and cosine waves in the samplers 226 and 227 produces

voltage outputs

228 and 229 representing the sine of 0 p and the cosine of 0- p, respectively. These voltages are delivered to holding

capacitors

236 and 231, respectively.

The equations set forth in the general theory of bone generation indicate that voltages representing the sine and cosine of and the sine and cosine of p are also needed. To obtain these voltages, a conductor 235 connected to the output 196 of the operational amplifier 137 carries the voltage representing the angle 0 (or its 180 counterpart) to an operational amplifier 236, the output 237 of which is is fed to a delay monostable multivibrator 238. The input to the multivibrator 23S is the square wave input 208 which flips the multivibrator to its quasi-stable state, and the magnitude of the input voltage 237 determines the duration of the unstable state. The output 239l from the delay multivibrator 238 triggers a narrow pulse generator 241i, the narrow pulse output 241 of which is fed to a pair of sampler gates 242 and 241.13. The input to the gate 242 is the sine wave 14 and the input to the gate 243 is the cosine wave 16. When the narrow pulse 241 opens the gates 2412 and 243 they sample the sine and cosine Waves and deliver their outputs and 245 to holding

capacitors

246 and 247, respectively. The voltages stored in these capacitors 2/6 and 247 represent the sine and cosine of the angle 6.

The output voltage 195 from the operational amplifier 193 is carried by a conductor 250 to an operational amplifier 251, the output 252 of which is delivered to a delay monostable multivibrator 253. The multivibrator 253 has the square wave input 208 and has an output 254 occurring at a time determined by the magnitude of the DC input 252. The pulse 254 triggers a narrow pulse generator 255. The output 256 from the narrow pulse generator is delivered to a pair of

sampler gates

257 and 258 one of which has the sine wave input 14 and the other of which has the cosine wave input 16. The output 259 from the sampler gate 257 is a voltage representing the sine qb and is delivered to a holding capacitor 260. The output 261 from the sampler gate S is a voltage representing the cosine q and is delivered to a holding capacitor 262.

From the foregoing it is evident that the holding

capacitors

217, 219, 230, 231, 246, 247, 266 and 262 store voltages representing the sine (6H- 15), cosine (o-t-qb) sine (t9-qi), cosine (-q), sine 0, cosine 0, sine qi), and cosine qt. These holding capacitors actually receive a number of sampled voltages each representing the appropriate sine or cosine function, because each sampler gate is opened a number of times during the generation of a bone. For example, the delay multivibrator 207 is dipped each time it receives a square wave input and therefore delivers successive narrow pulse outputs to the narrow pulse generator 216. The series of narrow straight sided pulses at the output of the generator 210 cause successive sarnplings of the sine and cosine waves in the

sampler gates

214 and 215 with these sample voltages being delivered to the holding

capacitors

217 and 219. Normally, these holding capacitors may receive about l5 to 2() sampled pulses during the generation of a bone.

There is a buffer amplifier 263 on the output side of each holding

capacitor

217, 219, 231i, 231, 246, 247, 260 and 262. The amplifiers 263 present a high output impedance to the holding capacitors, allowing the capacitors to hold accurate, unrippled, sampled voltages.

These sine and cosine functions could be generated in other Ways. For example, the inputs to the 0

gates

69, 77, etc., could be DC values previously resolved into sinecosine values by potentiometers, requiring, however, another row of 0 gates. Similarly the p gates 76, 7S etc., could have sine and cosine inputs. Any appropriate sinecosine function generator may be used.

Bone integrators To get quantities representing the X, Y and Z components of a bone being drawn, there are an X integrator 265, a Y integrator 266, and a Z integrator 267 shown in FlGURE 1. The X integrator 265 comprises a high gain amplifier 268 with a feed back capacitor 269 connected across it. The Y integrator 266 comprises a high gain amplifier 270 with a feed back capacitor 271 connected across it. The Z integrator at 267 comprises a high gain amplifier 272 with a feed back capacitor 273 connected across it. The input to the X integrator 265 includes the voltage representing the quantity cos (t2-Hp) from the holding capacitor 219 through the amplifier 263 delivered by a conductor 275 through a resistor 276 to an input conductor 277; and a voltage representing the quantity cos (0-) from the holding capacitor 231 carried by a conductor 278 through a resistor 279 to the input conduetor 277. The quantities cos (5H-qs) and cos (t9-rp) are halved and added by the

resistors

276 and 279, and the sum is presented to the input 277 of the integrator 265. From the equations set forth in the general theory of bone generation, the trigonometric equivalent to this sum is the quantity cos 0 cos e. Since the input 277 to the integrator 265 is a DC voltage, the output 280 from the integrator 265 is a ramp function representing the quantity klt cos 0 cos gb wherein k1 is a constant determined by the

resistors

276 and 279 and the capacitor 269 and t is the time variable. The charge on the feed back capacitor 269 determines the starting point of the ramp function klt cos 0 cos e, which starting point will be coincident with the ending point of the previous output 289 so long as the capacitor 269 is not discharged. Thus, unless the capacitor 26g is discharged, successive bones are joined together end to end as they are drawn or generated.

The input to the Y integrator 266 includes the voltage representing the quantity sin (IH-(p) delivered from the holding capacitor 217 by a conductor 28S through a resistor 286 to an input conductor 237 of the amplifier 27); and the Voltage representing the quantity sin (c-e) delivered from the holding capacitor 230 by a conductor 288 through a resistor 289 to the input conductor 287. Thus, the quantities sin (6H-p) and sin (-b) are halved and added together and presented to the integrator 266, but this input is equivalent to the quantity sin 0 cos 15. The ouput 29) from the integrator is a ramp function representing the quantity klt sin 0 cos qb. The starting point of the output 291) is determined by the presence or absence of a charge on the feed back capacitor 271.

The input to the Z integrator 267 is a voltage representing the quantity sin Q5 which is delivered from the holding capacitor 260 by a conductor 292 through a resistor 293 to the integrator amplifier 272. The output FlGURE 7 and its purpose is to scan a film 341, shown in FIGURE 9, to obtain a varying voltage, the instantaneous value of which represents the magnitude of the vector A. Thus, the magnitude of the vector A may be continuously changing as the vector twirls around a bone, and the purpose of the scanner 3d@ is to produce an output voltage that varies in proportion to the changes in length of the vector A.

A typica-l film 341 to be scanned might be divided into

sections

342, 343, 344 and 345 as shown in FIGURE 9. Each section is characterized by variations in density representing 360 or more of skin around the bones of various parts of a figure. These variations in density are proportional to the incremental lengths of the vector A `for an arm in section 342, a leg in section 343, the chest, neck and head in section 3M, and the hip in section 345.

`Referring to lFIGURE 7, the film 341 is placed in a film holder 347, positioned between a cathode ray tube 348 and a photomultiplier tube 349. There are appropriate lenses including an object lens `350 in front of the cathode ray tube 348, and condensing lenses 351 in front of the photomultiplier tube 349. When properly programmed, the .beam of the cathode ray tube 348 scans the `film 341 and valying intensities of the beam are focused through the condensing lenses 351 to the photomultiplier tube 349. The variations in intensity of the beam directed to the photomultiplier tube 349 are in proportion to the varying density of the material being scanned. The output 352 from the photomultiplier tube 3ft-9 is transmitted to a video amplifier i353 `whose output 354 is a voltage varying in amplitude in proportion to the varying intensity of the beam focused on the photomultiplier tube 349.

To scan the entire -film 341, the beam of the cathode ray tube 348 must be made to sweep in a horizontal direction and move in a vertical direction, or vice versa. ln 'the example shown in the drawings, the beam is caused to sweep in a horizontal direction by a horizontal defiection generator 359 that includes a sawtooth generator 360. The sawtooth generator comprises an operational amplifier 361 with a capacitor 362 connected across it. A switch 363 is connected in parallel with the capacitor 362 to ydischarge the capacitor when the switch is closed.

The amplifier 361 has a positive or a negative D.C. voltage input generated by a right-left bistable multivibrator 3&4. The magnitude of this voltage is variable according to the setting of a potentiometer 365. The function of the `bistable multivibrator 364 is to change the DC. voltage from positive to negative or back to positive, according to whether the direction of the horizontal sweep should be from right to left or from left to right. The direction of the sweep is dictated by whether a right or left appendage is being drawn. This is important when drawing the arms or the legs because the film 341 contains only one arm section 342 and one leg section 343 used for both the right and left arms and legs of the figure. Therefore, when scanning the skin for the left arm, the scanner must scan in opposite direction that it scans when skin is fbeing applied to the right arm. The same is true for the leg. On the other hand, it does not matter in which direction the scanning takes place for the single head, neck and chest, or for the single hips. However, this method is used to save film space. The -film would have separate sections of variable density for each appendage of a figure, as in the case of Captain Hook or Peg Leg Pete. The scanner would be programmed accordingly.

The "bistable multivibrator 3645 has two inputs, 36o and 367. As shown in FIGURE 6, one of these inputs, 3de, is -connected through a diode 363 to the conductor l2 which, in turn, is the input to the step counter group 325 for the right arm. Therefore, when a pulse is delivered to this storage counter group 325, the multivibrator 364 is caused to fiip to its `first condition, say a condition that generates a positive D.C. output. The multivibrator 354 remains in this first condition during the operation of the

storage counter groups

325, 326, 327 and 333 because during this time, there is no voltage input through the conductor 337. However, when a voltage is transmitted through a conductor 369 yfrom the storage counter group 328 to the first one of the storage counter group 329 to draw the left arm, a voltage is also transmitted through a conductor 376, a diode 371, and the conductor 367 to flip the multivibrator 364i to its second position. `ln this second condition, the output `from the multivibrator 364i is a D.C. voltage of the opposite, or negative, polarity. Themultivibrator 364 then remains in that condition during the period o-f oper-ation of the

storage counter groups

329, 330, 331 and 332.

The storage counter group 332 is connected to the group 333 lby a conductor 374. This conductor 374 and the conductor 369 between the

storage counter groups

328 and 329, are shown with releasable plug ends to indicate that changes can be made in the sequence of operation of the storage counter groups. Actually, the order of operation of these storage counters should ordinarily coincide with the positions of the bones, bones nearer the viewer being drawn before bones behind them.

Although the sweep of the cathode ray beam in the tube 3458 may be in either horizontal direction when scanning the chest, neck and head and the hips, the connection 374 illustrated causes the sweep to be in the same direction as the sweep for the right storage counters 325, 32d, 327 and 328 because there is a conductor 375 connected from the input 366 of the right-left multivibrator .3o/i through a diode 376 to the input side of the storage counter group 333. This conductor 375 delivers a pulse to the multivibrator 364 to flip the multivibrator back to its first or positive output condition when the storage counter group 333 receives a pulse.

The input to the sawtooth generator 360, which is always either a positive or a negative DC voltage of constant amplitude coming from the multivibrator 364, causes a gradual buildup of charge on the capacitor 362, thereby generating the sloping portion of the sawtooth wave, as is known in the art.

The switch 363 is normally open, but closes every time it receives a signal through its input 380. The input 380 is connected to the output of a bistable multivibrator 381, and the input 382 to the bistable multivibrator 381 is connected to the output of a delay monostable multivibrator 333.

The input to the delay multivibrator 333 is connected to the square wave output 12 from the master oscillator lo.

The -beginning of each square wave pulse causes the delay multivibrator 333 to iiip to its quasi-stable state. The input 332 to the bistable multivibrator 381 occurs at the time that the delay multivibrator flips back from its quasi-stable state to its stable state. Therefore, the time that the bistable multivibrator 381 changes Iits state depends upon the delay time of the delay multivibrator 383. This delay time is controlled by another input 384 to the delay multivibrator 383. The conductor 334 is connected through an adder 385 to the output side of an integrator that comprises an operational amplifier 386 with a capacitor 337 connected across it and a normally open switch connected in parallel with the capacitor 387 to discharge the capacitor when the switch 388 is closed. The amplitier 336 has an input 389, and to understand the nature of that input, FiGURE 2 must be reexamined.

Returning to FIGURE 2 and referring particularly to the

gates

71, 79, 87, and 149N, these are r gates wherein l' is a symbol representing the angular position of the skin about the bone. In other words, if a bone, like an arm bone, is turned, the value of r changes.

These r gates have variable inputs 39), 391, 392, 393 and 3915i for establishing different voltage values according to the different values of r for each bone. When the

Claims

1. A METHOD OF PRODUCING AN ANIMATED DISPLAY COMPRISING THE STEPS OF GENERATING A SERIES OF REFERENCE VOLTAGES, SUCH VOLTAGE HAVING A CHARACTERISTIC REPRESENTING LINES ON A DISPLAY SCOPE, EACH OF PREDETERMINED LENGTH ACCORDING TO THE LENGTHS OF PARTS OF AN OBJECT TO BE DISPLAYED, FOR EACH REFERENCE LINE VOLTAGE APPLYING SELECTED VOLTAGES TO THE DEFLECTION PLATES OF THE DISPLAY SCOPE CORRESPONDING TO THE POSITIONS OF THE REFERENCE LINES, GENERATING A VECTOR VOLTAGE, AND CONTINUOUSLY ADDING THE VECTOR VOLTAGE TO THE REFERENCE LINE VOLTAGES AS THEY ARE DRAWN ON THE DISPLAY TUBE.

8. A SYSTEM FOR PRODUCING ELECTRICAL INFORMATION FOR USE IN VIDEO READOUT TO DISPLAY AN ANIMATED MULTI-PART FIGURE ON THE FACE OF A DISPLAY TUBE COMPRISING A CLOCK, A PLURALITY OF CONTROLS SUCCESSIVELY OPERABLE BY THE CLOCK, EACH FOR A PRESELECTED TIME DURATION FOR GENERATING FIRST VOLTAGES, THE TIME DURATION OF EACH FIRST VOLTAGE BEING PROPORTIONED TO THE LENGTH OF A PART OF THE FIGURE, A PLURALITY OF POSITION PARAMETER NETWORKS EACH OPERATED BY A FIRST VOLTAGE, EACH POSITION PARAMETER NETWORK HAVING MEANS TO GENERATE A VOLTAGE FOR THE DURATION OF OPERATION OF THE CONTROL, THE VOLTAGES FROM THE POSITION PARAMETER NETWORKS BEING PROPORTIONED TO THE POSITIONS OF THE PARTS OF THE FIGURE WITH RESPECT TO PREDETERMINED REFERENCE COORDINATES, A GENERATING NETWORK INCLUDING MEANS TO GENERATE A VECTOR VOLTAGE AND MEANS TO MODULATE THE AMPLITUDE OF THE VECTOR VOLTAGE, A VOLTAGE COMBINING NETWORK FOR ADDING THE MODULATED VECTOR VOLTAGES TO THE FIRST VOLTAGES.